Research on Remote Monitoring and Fault Diagnosis System of New Energy Vehicles Based on Internet of Things

This research innovatively builds an intelligent monitoring and fault identification architecture based on Internet of Things technology, specifically targeting remote monitoring and real-time fault analysis of new energy vehicles. This architecture cleverly utilizes the advanced features of the Internet of Things to capture real-time data on the operating status of new energy vehicles, and seamlessly transmits the collected data to the central server for in-depth analysis and processing through wireless communication technology. In order to cope with the complex and changeable fault identification challenges, this study introduced two advanced machine learning models, decision tree and random forest, and significantly improved the accuracy of fault diagnosis and system response speed through data-driven strategies. This research covers key aspects such as data acquisition, preprocessing, feature extraction and fault classification, and innovatively integrates remote monitoring and fault warning mechanisms. Through model verification based on actual operating data, the simulation results highlighted that the random forest algorithm achieved an eye-catching fault diagnosis accuracy of $96.7 \%$, which was significantly better than the $89.5 \%$ of the decision tree algorithm. In addition, the significant shortening of system response time ensures its efficient application in actual scenarios and meets acceptable performance standards.